An Introduction to the Study of Experimental Medicine

At first sight, and considering things in a general way, this dis^ tinction between the experimenter's activity and the observer's passivity seems plain and easy to establish. But as soon as we come down to experimental practice we find that, in many instances, the separation is very hard to make, and that it sometimes even involves obscurity. This comes, it seems to me, from confusing the art of investigation, which seeks and establishes facts, with the art of reasoning, which works them up logically in the search for truth. Now in investigation there may be activity, at once of the mind and of the senses, whether in making observations or in making experiments.

Indeed, if we chose to admit that observation is characterized by this alone, that men of science note phenomena which nature produces spontaneously and without interference by them, still we could not conclude that the mind, like the hand, always remains inactive in observation ; and we should be led to distin^ish under this head two kinds of observations, some passive, others active. I assume, for instance, what often occurs, — that some endemic disease appears in a region and presents itself to a physician's observation. Here is a spontaneous or passive observation which the physician makes by chance and without being led to it by any preconceived idea. But after observing the first case, if the physician has an idea that the appearance of this disease may well be related to certain special meteorological or hygienic circumstances, he takes a journey to other regions where the same disease prevails, to see whether it develops under the same conditions. This second observation, made in view of a preconceived idea of the nature and cause of the disease, is what we must obviously call an induced or active observation. I should say as much of an astronomer who, in watching the sky, discovers a planet passing, by chance, before his telescope ; in this case he makes a fortuitous or passive observation, i.e., without a preconceived idea. But, if the astronomer, after noting the aberrations of a planet, goes on to make observations, to seek a reason for them, then I should say that he makes active observations, i.e., observations produced by a preconceived idea of the cause of the aberration. We might multiply instances of this kind ad infinitum, to prove that, in noting natural phenomena that present themselves, the mind is now passive, now active, — which means, in other words, that observations are made, now without a preconceived idea and by chance, and again with a preconceived idea, i.e., with intention to verify the accuracy of a mental conception.

On the other hand, if we concede, as we said above, that experiment is characterized by this alone, that men of science note phenomena which they have produced artificially and which would not naturally have presented themselves, even then we could not find that the experimenter's hand always actively interfered to bring about the appearance of these phenomena. In certain cases indeed we have seen accidents where nature acted for him ; and here again, from the point of view of manual intervention, we shall be forced to distinguish between active experiments and passive experiments. Let me assume that a physiologist wishes to study digestion and to learn what happens in a living animal's stomach; he will divide the walls of the abdomen and stomach according to known operative

rules and will establish what is called a gastric fistula. The physiologist will certainly think that he has made an experiment, because he has interfered actively to make phenomena appear which did not present themselves naturally to his eyes. But now, let me ask, did Dr. W. Beaumont make an experiment when he came across that young Canadian hunter who had received a point-blank gun-shot in the left hypochondria, and who had a wide fistula of the stomach in the scar, through which one could look inside that organ? Dr. Beaumont took this man into his service and was able to study the phenomena of gastric digestion de vim for several years, as he shows in the interesting journal which he has given us on this subject.^ In the first case, the physiologist acted on the preconceived idea of studying digestive phenomena and made an active experiment. In the second case, an accident produced a fistula of the stomach, and it presented itself fortuitously to Dr. Beaumont. According to our definition, he made a passive experiment. These examples therefore prove that, in verifying the phenomena called experiments, the experimenter's manual activity does not always come in, since it happens that the phenomena, as we have seen, may present themselves as fortuitous or passive observations.

But certain physiologists and physicians characterize observation and experiment somewhat differently. For them, observation consists in noting everything normal and regular. It matters little whether the investigator has produced the appearance of the phe^ nomena himself or by another's hands or by accident; he considers them without disturbing them in their natural state and so mak6s an observation. Thus, according to these authors, observations were made in both examples of gastric fistula cited above, because in both cases we had under our eyes digestive phenomena in their natural state. The fistula served only for seeing better and making observations under the most favorable conditions.

Experiment, according to the same physiologists, implies, on the contrary, the idea of a variation or disturbance that an investigatoi: brings into the conditions of natural phenomena. This definition corresponds, in fact, to a large group of experiments made in physiology, which might be called experiments by destruction. This form of experimenting, which goes back to Galen, is the simplest; it ■W. Beaumont, Experiments and Observations on the Oastrio Juice and on Physiological Digestion. Boston, 1834.

should suggest itself to the minds of anatomists wishing to learn, in the living subject, the use of parts that they have isolated by dissection in the cadaver. To do this, we suppress an organ in the living subject, by a section or ablation ; and from the disturbance produced in the whole organism or in a special function, we deduce the function of the missing organ. This essentially analytic, experimental method is put in practice every day in physiology. For instance, anatomy had taught us that two principal nerves diverge in the face : the facial (seventh cranial) and the trigeminal (fifth cranial) ; to learn their functions, they were cut, one at a time. The result showed that section of the facial nerve brings about loss of movement, and section of the trigeminal, loss of sensation, from which it was concluded that the facial is the motor nerve of the face, and the trigeminal the sensory nerve.

We said that, in studying digestion by means of a fistula, we merely make an observation, according to the definition which we are examining. But after we have established the fistula, if we go on to cut the nerves of the stomach, in order to se^e the changes which result in the digestive function, then, according to the same way of thinking, we make an experiment, because we seek to learn the function of a part from the disturbance which its suppression involves. And this may be summed up by saying that in experimentation we make judgments by comparing two facts, one normal, the other abnormal.

This definition of experiment necessarily assumes that experimenters must be able to touch the body on which they wish to act, whether by destroying it or by altering it, so as to learn the part which it plays in the phenomena of nature. As we shall later see, it is on this very possibility of acting, or not acting, on a body that the distinction will exclusively rest between sciences called sciences of observation and sciences called experimental.

But if the definition of experiment which we have just given differs from the definition examined in the first place in that it admits that we make an experiment only when we can vary or can dissociate phenomena by a kind of analysis, still it resembles the first in that it also always assumes an intentional activity on the experimenter's part, in producing a disturbance of the phenomena. Now it will be easy to show that the operator's intentional action can often be replaced by an accident. Here too, as in the first definition, we

might distinguish between disturbances occurring intentionally and disturbances occurring spontaneously or unintentionally. Indeed taking again the example in which a physiologist cuts the facial nerve to learn its function, I assume that a ball, a sabre cut or a splinter of stone, has cut or destroyed the facial nerve; there will result fortuitously a paralysis of movement, i.e., a disturbance, exactly the same as that which the physiologist caused intentionally.

It is the same in the case of numberless pathological lesions which are real experiments, by which physicians and physiologists profit, without any purpose on their part to produce the lesions, which result from disease. I emphasize this idea now, because it will be useful to us later, to prove that medicine includes real experiments which are spontaneous, and not produced by physicians.^ I will make one more remark by way of conclusion. If indeed we characterize experiment by a variation or disturbance brought into a phenomenon, it is only in so far as we imply that the disturbance must be compared with the normal state. As experiments indeed are only judgments, they necessarily require comparison between two things; and the intentional or active element in an experiment is really the comparison which the mind intends to make. Now, whether the alteration is produced by accident or otherwise, the experimenter's mind compares none the less. It is therefore unnecessary to regard as a disturbance one of the facts to be compared, especially as there is nothing disturbed or abnormal in nature ; everything happens according to laws which are absolute, i.e., always normal and determined. Effects vary with the conditions which bring them to pass, but laws do not vary. Physiological and pathological states are ruled by the same forces ; they differ only because of the special conditions under which the vital laws manifest themselves.

II. Gaining Experience and Relying on Observation Is Different FROM Making Experiments and Making Observations The general objection which I make to the preceding definitions is that they give words too narrow a meaning, by taking account of only the art of investigation, instead of considering observation and experiment at the same time as the two opposite extremes of experi- 'Lallemand, Propositions de pathologic tendant d iclairer plusfieurs points de physiologie. Thesis. Paris, 1818. 2nd edition, 1824.

mental reasoning. So we find these definitions lacking in clearness and generality. To give the definition its full usefulness and value, therefore, I think that we must distinguish what pertains to the method of investigation, used to gather facts, from the characteristics of the intellectual method, which utilizes facts and makes them at once the support and the criterion of the experimental method. In French the word experience in the singular means, in general and in the abstract, the knowledge gained in the practice of life. When we apply to a physician the word experience in the singular, it means the information which he has gained in the practice of medicine. It is the same with the other professions; and it is in this sense that we say that a man has gained experience, or that he has experience. Subsequently the word experience (experiment) in the concrete was extended to cover the facts which give us experimental information about things.

The word observation in the singular, in its general and abstract use, means noting a fact accurately with the help of appropriate studies and means of investigation. In the concrete the word observation has been extended to cover the facts noted; and it is in this sense that we speak of medical observations, astronomical observations, etc. Speaking concretely, when we say "making experiments or making observations," we mean that we devote ourselves to investigation and to research, that we make attempts and trials in order to gain facts from which the mind, through reasoning, may draw knowledge or instruction.

Speaking in the abstract, when we say "relying on observation and gaining experience," we mean that observation is the mind's support in reasoning, and experience the mind's support in deciding, or still better, the fruit of exact reasoning applied to the interpretation of facts. It follows from this that we can gain experience without making experiments, solely by reasoning appropriately about wellestablished facts, just as we can make experiments and observations without gaining experience, if we limit ourselves to noting facts.

Observation, then, is what shows facts; experiment is what teaches about facts and gives experience in relation to anything. But as this teaching can come through comparison and judgment only, i.e., by sequence of reasoning, it follows that man alone is capable of gaining experience and perfecting himself by it. "Experience," says Goethe, "disciplines man every day." But this is because man reasons accurately and experimentally about what he observes ; otherwise he could not correct himself. The insane, who have lost their reason, no longer learn from experience; they no longer reason experimentally. Experience, then, is the privilege of reason. "Only man may verify his thoughts and set them in order; only man may correct, rectify, improve, perfect and so make himself every day more skilful, wise and fortunate. Finally for man alone does the art exist, that supreme art of which the most vaunted arts are mere tools and raw material: the art of reason, reasoning." *

In experimental medicine, we shall use the word experience in the same general sense in which it is still everywhere used. Men of science learn every day from experience ; by experience they constantly correct their scientific ideas, their theories; rectify them, bring them into harmony with more and more facts, and so come nearer and nearer to the truth. We can learn, — i.e., gain experience of our surroundings, — in two ways, empirically and experimentally. First there is a sort of teaching or unconscious and empirical experience, which we get from dealing with separate objects. But the knowledge which we gain in this way is also accompanied necessarily by vague experimental reasoning which we carry on quite unawares, and in consequence of which we bring together facts to make a judgment about them. Ex- perience, then, may be gained by empirical and unconscious reasoning; but the obscure and spontaneous movement of the mind has been raised by men of science into a clear and reasoned method, which therefore proceeds consciously and more swiftly toward a definite goal. Such is the experimental method in the sciences by which experience is always gained by virtue of precise reasoning based on an idea born of observation and controlled by experiment. In all experimental knowledge, indeed, there are three phases; an observation made, a comparison established and a judgment rendered. By the experimental method, we simply make a judgment on the facts around us, by help of a criterion which is itself just another fact so arranged as to control the judgment and to afford experience. Taken in this general sense, experience is the one source of human knowledge. The mind in itself has only the feeling

of a necessary relation between things : it can know the form of that relation only by experience. Two things must, therefore, be considered in the experimental method: (1) The art of getting accurate facts by means of rigorous investigation; (2) the art of working them up by means of experimental reasoning, so as to deduce knowledge of the law of phenomena. We said that experimental reasoning always and necessarily deals with two facts at a time : observation, used as a starting point ; experiment, used as conclusion or control. Tn reasoning, however, we can distinguish between actual observation and experiment only, as it were, by logical abstraction and because of the position in which they stand.

But outside of experimental reasoning, observation and experiment no longer exist in this abstract sense; there are only concrete facts in each, to be got by precise and rigorous methods of investigation. We shall see, further on, that the investigator himself must be analyzed into observer and experimenter; not according to whether he is active or passive in producing phenomena, but according to whether he acts on them or not, to make himself their master.

The art of investigation is the cornerstone of all the experimental sciences. If the facts used as a basis for reasoning are ill-established or erroneous, everything will crumble or be falsified; and it is thus that errors in scientific theories most often originate in errors of fact. In investigation, considered as the art of experimental research, we find only facts brought to light by investigators and noted as rigorously as possible with the help of the most suitable means. There is no further occasion here to distinguish observers from experimenters by the character of the processes of investigation used. In the last section I showed that the definitions and distinctions which men have tried to set up on the basis of the investigator's activity or passivity cannot be sustained. Observers and experimenters, indeed, are investigators seeking to note facts to the best of their ability, using more or less complicated means for this purpose according to the complexity of the phenomena that they study. Both need the same manual and intellectual activity, the same

dexterity, tiie same spirit of invention, to create and perfect the different pieces of apparatus or instruments for investigation which, for the most part, they have in common. Every science has its own kind of investigation and its equipment of special instruments and methods. This, after all, is plain enough, since every science is characterized by the nature of its probleriis and by the variety of the phenomena that it studies. Medical investigation is the most complicated of all: it includes all the methods proper to anatomical, physiological and therapeutic research, and, as it develops, it also borrows from chemistry and physics many means of research which become powerful allies. In the experimental sciences all progress is measured by improvement in the means of investigation. The whole future of experimental medicine depends on creating a method of research which may be applied fruitfully to the study of vital phenomena, whether in a normal or abnormal state. I shall not here dwell on the necessity of such a method of investigation in experimental medicine, and I shall not even attempt to enumerate the difficulties. I shall limit myself to saying that my whole scientific life is devoted to contributing my share to the immense work which modern science will have the glory of having understood, and the merit of having begun, while leaving to future ages the task of continuing and finally establishing it. The two volumes which will form my work on the Principles of Experimental Medicine will be devoted solely to elaborating the methods of experimental investigation applied to physiology, pathology and therapeutics. But as no one man can consider all aspects of medical investigation, I shall limit myself further in this vast subject, by dealing especially with systematization of the methods of zoological vivisection.

It cannot be gainsaid that this is the most delicate and difficult branch of biological investigation ; but I deem it the most fruitful and perhaps the most immediately useful for the advancement of experimental medicine. In scientific investigation, minutiae of method are of the highest importance. The happy choice of an animal, an instrument constructed in some special way, one reagent used instead of another, may often suffice to solve the most abstract and lofty questions. Every time that a new and reliable means of experimental analysis makes its appearance, we invariably see science make progress in the questions to which this means of analysis can be applied. On

the contrary, a bad method or defective processes of research may cause the gravest errors, and may retard science by leading it astray. In a word, the greatest scientific truths are rooted in details of experimental investigation which form, as it were, the soil in which these truths develop. One must be brought up in laboratories and live in them, to appreciate the full importance of all the details of procedure in investigation, which are so often neglected or despised by the false men of science calling themselves generalizers. Yet we shall reach really fruitful and luminous generalizations about vital phenomena only in so far as we ourselves experiment and, in hospitals, amphitheatres, or laboratories, stir the fetid or throbbing ground of life. It has somewhere been said that true science is like a flowering and delectable plateau which can be attained only after climbing craggy steeps and scratching one's legs against branches and brushwood. If a comparison were required to express my idea of the science of life, I should say that it is a superb and dazzlingly lighted hall which may be reached only by passing through a long and ghastly kitchen.

IV. Obseevers and Experimenters; the Sciences of Observation AND OF Experiment We have just seen that, from the point of view of the art of investigation, observation and experiment should be considered only as facts brought out by investigators, and we have added that methods of investigation do not differentiate the men who observe from the men who experiment. Wliere then, you will ask, is the difference between observers and experimenters ? It is here : we give the name observer to the man who applies methods of investigation, whether simple or complex, to the study of phenomena which he does not vary and which he therefore gathers as nature offers them. We give the name experimenter to the man who applies methods of investigation, whether simple or complex, so as to make natural phenomena vary, or so as to alter them with some purpose or other, and to make them present themselves in circumstances or conditions in which nature does not show them. In this sense, observation ia investigation of a natural phenomenon, and experiment is investigation of a phenomenon altered by the investigator. We shall see that this distinction, apparently quite external and depending simply on a

definition of words, still supplies the one meaning with which to grasp the important difference separating sciences of observation from sciences of experimentation or experimental sciences. We said, in an earlier paragxaph, that the words observation and experiment, taken in an abstract sense, mean, the first, purely and simply noting a fact, the second, testing an idea by a fact But if we consider observation merely in this abstract sense, we cannot deduce from it any science of observation. By simply noting facts, we can never succeed in establishing a science. Pile up facts or observations as we may, we shall be none the wiser. To leam, we must necessarily reason about what we have observed, compare the facts and judge them by other facts used as controls. But one observation may serve as control for another observation, so that a science of observation is simply a science made up of observations, i.e., a science in which we reason about facts observed in their natural state, as we have already defined them. An experimental science, or science of experimentation, is a science made up of experiments, i.e., one in which we reason on experimental facts found in conditions created and determined by the experimenter himself.

Certain sciences, like astronomy, will always remain sciences of observation, because the phenomena studied are outside our sphere of action ; but terrestrial sciences may be, at once, sciences of observation and experimental sciences. Let me add that all these sciences begin as sciences of pure observation; only as we go into the analysis of phenomena do they become experimental, because the observer, turning experimenter, invents methods of investigation to penetrate bodies and vary the conditions of phenomena. Experimentation is only utilizing methods of investigation peculiar to experimenters.

Now experimental reasoning is absolutely the same, whether in sciences of observation or in experimental sciences. We find the same judgment by comparison based on two facts, one used as starting point, the other as conclusion, of our reasoning. Only in the sciences of observation, the two facts are always observations ; while in the experimental sciences, the two facts may be taken exclusively from experimentation, or at the same time from experimentation and from observation, according to the special case and according to how deeply we go into experimental analysis. A physician observing a disease in different circumstances, reasoning about the influence of these

circumstances, and deducing consequences which are controlled by other observations, — this physician reasons experimentally, even though he makes no experiments. But if he wishes to go further, and to know the inner mechanism of the disease, he will have to deal with hidden phenomena, and so he will experiment ; but he will still reason in the same way. A naturalist observing animals in all the conditions necessary to their existence, and deducing from these observations consequences verified and controlled by other observations, — such a naturalist uses the experimental method even though he performs no experiments, properly speaking. But if he has to go on to observe phenomena inside the stomach, he is forced to invent more or less complex methods of experimentation in order to look inside a cavity hidden from sight. His experimental reasoning, nevertheless, is the same; Reaumur and Spallanzani alike apply the experimental method when making their observations of natural history or their experiments with digestion. When Pascal made a barometric observation at the bottom of the Tour Saint Jacques, and later took another at the top of the tower, we must admit that he performed an experiment; yet here were simply two comparative observations of air pressure carried out in view of the preconceived idea that this pressure should vary according to height. On the other hand, when Jenner,^ in observing a cuckoo on a tree, used a spy-glass so as not to frighten it, he made a mere observation, because he did not compare this cuckoo with a previous cuckoo, to deduce a conclusion from the observation and to form a judgment about it. In the same way an astronomer first makes observations and then reasons about them to deduce a system of ideas which he controls by observations made in conditions suited to his purpose. The astronomer reasons like an experimenter, because the experience which he gains implies judgment throughout and comparison between two facts bound together in the mind by an idea.

However, as we have said already, we must clearly differentiate astronomers from the men of science concerned with terrestrial science, in that astronomers limit themselves perforce to observation, as they cannot go into the skies to experiment on the planets. In this power of the investigator to act on phenomena, precisely here • Jenner, On the natural history of the cuckoo. {Philosophical Transactions, 1788, Chap. XVI, p. 432.) is the difference separating the so-called sciences of experimentation from those of observation.